Moulton Laboratories
the art and science of sound

Why Recordings Don't Sound Quite Like the Real Thing and Some Things You Can Do About It. An informal introduction to the realities of psychoacoustics.

< 1 2 3 4 5 6 >

An Overview

The microphone is actually a fairly simple device, conceptually. It consists of a very delicate diaphragm suspended in air that moves back and forth due to air pressure changes (sound waves). That diaphragm is connected to one of several different electromagnetic mechanisms that converts the motion of the diaphragm to an alternating electrical current. That current flows in a cable connected to the mixing console, and becomes the basis for the audio signal that we will process, record, and ultimately send to a loudspeaker, where it will be converted back to sound. The process can be described in a fairly straightforward way: waves of air pressure buffet the diaphragm, which moves back and forth as a result. That motion is converted into an equivalent electrical motion. Voila! Audio is created!

The ear, on the other hand, is a very complicated device conceptually. It consists of a very delicate diaphragm suspended in air that moves back and forth due to air pressure changes (sound waves). That diaphragm is connected, via a fairly elaborate mechanical linkage, to a remarkable organ called the basilar membrane. At the basilar membrane, the mechanical motions are converted to neurological impulses that are sent to our brain. There, along with some other things, those impulses are presented to our conscious mind. Voila! We hear sound!

In a rudimentary way, then, the microphone converts sound into an analogous electrical waveform, while the ear converts it into neurological impulses. So what's the big deal? How does that difference affect our approach to the recording process?

The first thing you've got to understand might be called the paradox of perception: what we hear is not what comes in our ears. What we consciously hear is an extremely highly edited, polished, rewritten, equalized, compressed, noise-reduced, vocoded, positive-feedback-looped, delayed, stereo-enhanced, reorganized, remixed, mastered, converted to digital pulse-code construction that has very little relationship to the raw physical sound data that went into our ears.

A couple of years ago I had a medical misadventure which required that, for diagnostic testing, I take an anti-convulsant drug that was also, incidentally, psychoactive (which is high-tech slang for "mind-bending"). One of the systems it affected was the auditory system -- it tended to inhibit outbound neurological impulses sent from the auditory cortex in the brain to the basilar membrane in each ear. The result was a dramatic change in the amount of signal-processing that occurred to the inbound impulses of auditory data en route to my poor, confused brain. I noticed differences particularly in noise-based sounds. The hissing, rushing sound of a shower, for instance, turned into a rough, industrial-strength crackling, almost like a wood fire. Musical sounds seemed rough, poorly defined, and with very clangorous, ambiguous pitch. It was weird there, for a couple of days. And the insight I gained, after asking the doctor just what the fuck he was giving me (the drug makes you a little violent, too!) and was it really supposed to destroy my hearing (dark thoughts of malpractice suits zinging away through what remained of my nervous system!), was that what I have come to expect music to sound like isn't really what comes in my ears. My own auditory mechanism, when it isn't ripped out of its gourd (another high-tech expression for "a little confused right now"), adds a lot, an awful lot, to the process.
NEXT> Comparing Ear and Mic    
< 1 2 3 4 5 6 >

COMMENTS

Staten Island, NY     Jun 10, 2006 11:36 PM
These articles are amazing, genius.... i can't wait to read all of them and then re-read all of them again... - Christopher
Christopher Sauter 
USA     Mar 12, 2010 03:46 AM
The article seems to miss that whether a mic/speaker or direct sound is emanating, the ear is going to be used and all its cool features and the brain processes inherent in that, unless we are talking about robots enjoying music. The comparison should instead be the sound that enters the ear in each instance. You would not want a mic to do the processing the ear does because the processing would be doubled with, no doubt, strange results.
Also, even though light moves faster than sound the brain take more, not less time to process vision. Watching the other musician instead of listening is probably not the best choice unless you can perceive the changes more clearly with vision than sound.
I also seriously doubt that we can tell what direction sound is coming from with one ear other than by moving our heads or already knowing how loud a sound should be if we were aimed at it optimally.
I have some hearing loss in one ear. If I put a hearing plug that ear, that mutes it by an additional 33DB, I am pretty close to deaf in that ear in that situation (which I have to do some times because the muscle he was talking about goes nuts sometimes). When that is the case I truly can't tell what direction sound is coming from with my good ear. My guess is that you are picking up some small bit of sound in the covered ear that is providing you with its direction. One ear direction I say is a total myth. Memory of a sound as it changes typically going around ones head may give cues we pick up, but if there is no relative movement history of the sound, I say I highly doubt any direction can be discerned.
mindbreaker 
Sydney, Australia     Dec 13, 2011 09:42 PM
mindbreaker is quite right; this article shows a complete misunderstanding of the way the acoustic recording/playback system works.

We would want the microphone to be similar to the ear if we were going to bypass the ear of the listener and insert electric signals directly into his nervous system, but this is, of course, not what we intend.

The role of the microphone is to capture information that is 'encoded' in air pressure fluctuations and convert it into a forma that can be stored. The loudspeaker is then called upon to convert this information back into its original form. Of course, there are many problems with this system, mostly at the loudspeaker end, but what matters at the microphone end is the completeness of the information-capture, not the exact mechanism.

Suppose, in an analogous case, that we take a page of handwriting, scan it into a computer, and then print it back into a new but almost identical page. It makes no difference that the scanner works in a totally different way from that in which the eye does, we just want it to capture as much information as possible.
Tim Smith 
Groton, MA     Dec 14, 2011 09:44 AM
Without spending a lot of time on this, there IS other research suggesting how single-ear localization happens. Further, I was startled by how robust I found that localization to be, even subject to its limitations. I'm not able to comment on mindbreaker's personal experience. My experience remains anecdotal, but easily, if fuzzily, repeatable. Thanks for writing, mindbreaker!
Dave Moulton 
Groton, MA     Dec 18, 2011 12:24 PM
Tim Smith makes some really good points in his post (although I disagree about my completely misunderstanding how acoustic recording/playback works – actually, I think I DO understand it).

Anyway, in this article I was comparing the ear to the microphone to illuminate various things about each of them, and also to show some places where we typically get confused. Tim isn't confused – he's got a really good handle on it.

The problems he doesn't get to have to do with where errors accrue in the capture of information at the microphone. We lose a lot there. Take a look at:

www.moultonlabs.com/weblog/more/we_want_really_accurate_recordings

Anyway, thanks, Tim, for an excellent and thoughtful post.

Best regards,

Dave
Dave Moulton 

Post a Comment



rss2

rss atom